Autonomic resource allocation for video streaming services in Content Delivery Networks

Video streaming is currently one of the most popular online services. In order to improve video the Quality of Service (QoS) of video streaming, Content Delivery Networks (CDNs) have been widely adopted. CDNs replicate their content on distributed servers, meaning that a user can access content from the nearest CDN server. In order to provide better QoS for video streaming, it is necessary to understand the current state of network resources and efficiently allocate them to a service user. In this paper, we capture the current state of a service by observing low-level monitoring data using ontological reasoning. By taking into account both latency and bandwidth utilization, we propose bandwidth allocation and request routing algorithms to guarantee Service Level Agreement.